Hyperledger Iroha can be deployed in different ways, depending on the perspective and the purpose. There can be either a single node deployed, or multiple nodes running in several containers on a local machine or spread across the network — so pick any case you need. This page describes different scenarios and is intended to act as a how-to guide for users, primarily trying out Iroha for the first time.
Due to a known issue with gRPC it might not be safe to deploy Iroha in production natively on MacOS or on older versions of Linux.
The issue might cause Iroha to crash due to socket exhaustion and unless it is set up correctly, using, say, Docker, it will need to be restarted manually which might disrupt the work. You can learn more about why the issue affects MacOS and noncontemporary Linux systems in this article (the way gRPC uses polling engine in Mac is susceptible to this type of vulnerability and unlike Linux it cannot be avoided by using other options of polling engines).
So, to be on the safe side, please try to deploy on newer versions of Linux (see the version numbers in the article above) and use Docker with Linux if you prefer MacOS in the production environment.